The invention provides a train energy-saving auxiliary driving method and system based on safety reinforcement learning, can be applied to the technical field of artificial intelligence and Internet of Things, and particularly relates to the field of intelligent driving. The method comprises the following steps: inputting actual traction voltage and actual traction current of a train into a train operation model to obtain estimated energy consumption of the train; according to the estimated energy consumption and train communication and control data of the train, updating model parameters of the train operation model to obtain an updated train operation model; inputting train communication and control data and train line condition information and operation plan information into the updated train operation model, and calculating a recommended speed curve and a recommended working condition of a future operation process of the train in combination with a reinforcement learning algorithm; and performing energy-saving auxiliary driving on the train according to the recommended speed curve and the recommended working condition.
本公开提供了一种基于安全强化学习的列车节能辅助驾驶方法及系统,可以应用于人工智能、物联网技术领域,具体涉及智能驾驶领域。该方法包括:将列车的实际牵引电压和实际牵引电流输入列车运行模型,得到列车的估计能耗;根据估计能耗和列车的列车通控数据,对列车运行模型的模型参数进行更新,得到更新后列车运行模型;将列车通控数据和列车的线路条件信息、运行计划信息输入更新后列车运行模型,结合强化学习算法,计算列车的未来运行过程的推荐速度曲线和推荐工况;根据推荐速度曲线和推荐工况对列车进行节能辅助驾驶。
Train energy-saving auxiliary driving method and system based on safety reinforcement learning
基于安全强化学习的列车节能辅助驾驶方法及系统
2024-02-06
Patent
Electronic Resource
Chinese
IPC: | B61L Leiten des Eisenbahnverkehrs , GUIDING RAILWAY TRAFFIC |
Train driving safety auxiliary method and system based on train-ground cooperation
European Patent Office | 2024
|European Patent Office | 2024
|European Patent Office | 2024
|European Patent Office | 2025
|